We report on efficient laser operation of high quality crystalline Yb3+:Lu2O3 in thin disk configuration. Using doping concentrations between 1 at.% to 3 at.% and disk thicknesses between 0.08mm and 0.45mm the optimum crystal parameters have been determined. Pumped at 976 nm the laser operates at 1034 nm and 1080 nm. With a 0.25mm thick 3 at.% Yb:Lu2O3 disk 32.6W of output power at 45.3W incident pump power with a slope efficiency of 80% and a resulting optical-to-optical efficiency of 72% have been realized. These are the highest values in terms of slope efficiency as well as optical-to-optical efficiency for an Yb-doped thin disk laser reported so far. Using an 1mm birefringent filter continuous tuning from 987 nm to 1127 nm with more than 10Wof output power over a tuning range of 90 nm has been achieved.
© 2007 Optical Society of America
In the last decade the thin disk laser concept invented in 1993 by A. Giesen  entered into many fields of applications like material processing or the generation and amplification of ultrashort pulses. Due to the special design with its effective cooling mechanism an output power of more than 5 kW from one disk with high efficiency and beam quality has been achieved .
In recent years the research for new materials for the use in this setup has progressed rapidly. Although efficient thin disk laser operation has been demonstrated with Ytterbium doped tungstates [3, 4], borates , and vanadates , the well investigated Yttrium-Aluminum-Garnet (YAG)  is still the material of choice regarding high power applications in the kWrange. This is due to an excellent conjunction of high cross sections, high thermal conductivity, and the easy growth of laser crystals. Because of energy migration and transfer to impurities, a strong decrease in quantum efficiency towards higher Yb-concentrations is observed, which limits the usability of highly doped Yb:YAG as active media so far to a maximum doping level of about 10 at.%. However, high doping concentrations are needed to reduce the disk thickness and thus the maximum temperature and phase distortions, making fundamental mode operation easier to achieve with higher efficiencies.
Due to their excellent thermo-mechanical properties the sesquioxides Sc2O3, Y2O3, and Lu2O3 are attractive host materials for high power solid-state lasers [7–13]. Within these materials Lu2O3 provides the highest doping potential for Yb3+. Due to the very similar mass of Lu3+ and Yb3+, an increasing doping concentration induces nearly no scattering of the propagating phonons maintaining thus a high heat conductivity.
However, arising from their high melting temperatures in the range above 2400°C the production of sesqioxide bulk material in a satisfying quality has some difficulties. Due to not fully optimized growth conditions in previous growth experiments performed in our group, the cw and mode locked laser efficiency of bulk Yb:Lu2O3 was limited to 49% and 44%, respectively, so far [3, 10].
The fabrication of ceramic sesquioxides as an alternative to the difficult crystal growth has also been investigated intensely [14–23]. With ceramic Y2O3 a slope efficiency in cw operation of more than 80% with output powers in the W-range has been demonstrated . However, with ceramic Yb:Lu2O3, the relatively low efficiency of bulk material could not be surpassed so far. In cw and mode-locked laser operation, efficiencies of only 53% and 32%, respectively, have been demonstrated [15, 19].
In this work we report on improvement of cw laser operation of Yb:Lu2O3 using high purity, high optical quality crystals in the thin disk geometry under multi-pass pumping with diode lasers.
In Sect. 2 the spectroscopic and thermal properties of Yb:Lu2O3 are introduced. The experimental setup is described in Sect. 3, followed by the presentation and discussion of the experimental data. The main results of this work are summarized in the final section.
2. Thermal and spectroscopic properties of Yb:Lu2O3
One of the most desirable properties of host crystals for high power solid state lasers is a high thermal conductivity. From this point of view the sesquioxides Sc2O3, Y2O3, and Lu2O3 appear to be suitable due to their high initial heat conductivity in undoped crystals. However, like in all host materials the heat conductivity drops significantly when doped with Yb (Tab. 1). Due to the different atomic weights this effect is strongest in Sc2O3 and weakest in Lu2O3. Thus, even for higher doping levels the thermal conductivity of Lu2O3 is much larger than for Yb-doped YAG.
Figures 1(a) and 1(b) show the absorption and emission cross sections of Yb-doped Lu2O3 in comparison with Yb:YAG. In contrast to Yb:YAG for which the absorption cross sections σabs are similar around 940 nm and 969 nm, in Yb:Lu2O3 σabs is about 4 times larger for the zero-phonon line (ZL) around 976 nm than for the 950 nm absorption band. This allows on the one hand a strong reduction of the disk thickness for the same doping concentration leading to a better heat removal from the disk; on the other hand, with a FWHM of 2.1 nm the absorption bandwidth is considerably narrower than for the 950 nm band, which increases the demands on the bandwidth for the pump diodes and makes the laser more sensitive to shifts of the pump wavelength. Nevertheless, laser diodes fulfilling these demands are commercially available at output powers in the 100W range.
As can be seen in Fig. 1(b) the emission cross section of the main laser transition around 1034 nm is somewhat lower but slightly broader than for Yb:YAG. This results in a broader gain profile and enables the generation of shorter pulses with Yb:Lu2O3 (220 fs) than for Yb:YAG (340 fs), when the laser operates in a mode-locked regime [10, 25].
Additionally a second emission band at longer wavelength exists, which enables efficient laser operation also at 1080 nm and allows a broad tuning of the laser wavelength.
Previous crystals [3, 12] have been grown by the heat exchanger method (HEM) from Re-crucibles. To increase the crystal purity and quality, the crystals used in our experiments have been grown by the same technique but with improved growth conditions, that is optimized athmosphere using 6N Lu2O3 and 5N Yb2O3 in a high purity Re-crucible. Thus, polycrystalline boules of 40 mm diameter and 25 mm length with monocrystalline regions of up to 5 cm3 could be obtained. A more detailed discussion of the growth will be published elsewhere .
Because energy transfer to impurities leads to a decrease in quantum efficiency, the fluorescence lifetime can be used as an indicator for the crystal quality. Thus, to compare the purity of the new crystals with that of samples grown in former experiments in our group, the fluorescence lifetime has been measured. Using the pinhole-method [27, 28] to prevent radiation trapping values between 820 μs and 905 μs have been determined (Fig. 2). In contrast to previous growth experiments no concentration quenching occurred up to the highest fabricated doping concentration of 5 at.%, corresponding to an Yb-density of 14.5∙1020cm-3.
3. Experimental laser setup, results and discussion
The experimental laser setup is shown in Fig. 3. For the laser experiments crystals with doping concentrations of 1 at.%, 2 at.%, and 3 at.% and disk thicknesses between 0.08mm and 0.45mm have been prepared. The crystal diameter for all samples was 5mm.
One side of each disk had a broad high reflective (HR) coating from 900 nm to 1150 nm, including the pumping wavelength and the emission band. This side was soldered to a watercooled copper heat-sink by using an indium-tin metallic layer that decreased the thermal impedance between the laser medium and the cooling system. The opposite side of each crystal was anti reflective (AR) coated for the same spectral range. The fiber-coupled diode pump laser (JENOPTIC LaserDiode GmbH, Germany) delivered 45W at 976 nm with a spectral width of 2 nm; the fiber diameter was 600 μm with NA = 0.22. Efficient absorption of the pump radiation was achieved with a pumping module that provided 24 passes of the pump beam through the crystals. The pump beam diameter on the crystal was 1.2mm. A plane-concave resonator of ~70mm length was used to investigate the characteristics of the laser emission; the radius of curvature of the concave output mirror was 100mm. To achieve optimum absorption, the diode temperature was varied between 10°C and 21°C.
Figure 4 presents the output power versus the pump power (measured at the fiber end) for the most efficient crystal used in our experiments for different outcoupling mirrors. The 0.25mm thick 3 at.%-doped Yb:Lu2O3 crystal delivered at an optimum outcoupling transmission of TOC = 1.6% a maximum output power of 32.6W at a pump power of 45.3W; the maximum slope efficiency ηs and opt.-opt. efficiency ηopt was 80% and 72%, respectively. It can be seen in Fig. 4 that even for TOC = 0.4% the slope efficiency is well above 75%, which is an indicator for very weak crystal losses. As known from previous experiments , the laser wavelength changes from 1080 nm to 1034 nm at outcoupling rates exceeding a few percents. This can be explained by the higher inversion due to the increased laser threshold at higher outcoupling transmission, which shifts the wavelength of highest gain to 1034 nm. For that reason at TOC = 1.2%a slight decrease of the slope efficiency could be observed as the laser operated at 1080 nm and 1034 nm simultaneously.
The inset in Fig. 4 shows a typical laser emission spectrum of the 3 at.%-doped 0.25mm thick Yb:Lu2O3 disk. Even with a relatively low specified reflectivity of the AR-coating of about 0.02%, the laser emission is structured, due to an etalon effect between the front and the backside of the crystal. The wavelength separation between two laser modes is ~1 nm, which is in good accordance with the disk thickness of 0.25mm.
The beam quality M2, measured with a DataRay Beamscope-P7 beam analyzer was determined to be between 10 and 20 in this configuration.
To compare the laser performance of the different doping concentrations the slope efficiencies using the optimum outcoupling transmission of 1.6% of all examined crystals are shown in Fig. 5, where ηs is plotted versus an “effective disk thickness” Leff. For example, a 3 at.%- doped 0.1mm crystal has the same absorption as an 1 at.%-doped 0.3mm thick disk and is thus treated as a 0.3 at.%∙mm crystal. As can be seen in Fig. 5 ηs increases with the disk thickness and doping level. This can be attributed to the stronger absorption and weaker bleaching, simultaneously of the thicker disks. To confirm this interpretation a double pass pump experiment with the same resonator design was realized. Thus, it was possible to measure directly the absorbed pump power. The setup was tested with a 2 at.%-doped 0.12mm crystal, which had an effective disk thickness of 0.24 at.%∙mm and an ηs of 67% (cf. Fig. 5). Versus absorbed pump power the slope efficiency ηabs was measured to be about 80%. This is in good agreement with the disks, which absorb nearly 100% of the pump light in the TDL-setup.
As a result, it can be stated that the slope efficiency ηs is mainly determined by the maximum pump light absorption. No differences between the crystals of different growths runs have been observed. Furthermore, up to the maximum disk thiskness of 0.45mm no decrease in slope efficiency, caused by an increased thermal loading of the disk could be observed. This feature is due to the high thermal conductivity of Lu2O3 and a proof for an efficient heat removal even from thicker disks.
The tuning range of the laser emission was investigated using a 0.45mm 1 at.% and a 0.25mm 3 at.%Yb:Lu2O3 disk with different outcoupling mirrors under 32Wof incident pump power. The experiments were performed with a 1mm thick birefringent filter (BF) that was inserted under Brewsters angle into the laser resonator. To insert the BF the resonator length had to be increased to 150mm, using concave output couplers with 500mm radius of curvature.
With the 1 at.%-doped crystal and TOC = 1.6% laser emission could be achieved in two wavelength regions that is between 1001 nm and 1051 nm as well as 1067 nm and 1089 nm (Fig. 6). Reducing the outcoupling to 0.4% continuous tuning from 987 nm to 1098 nm with a maximum of 17.3W at 1034 nm was possible. Using a HR outcoupling mirror the total tuning range could even be extended to 126 nm from 1001 nm to 1127 nm; with more than 5W of output power over a tuning range of 90 nm, limited at the short wavelength side by the coating of the HR mirror.
Using the 3 at.%-doped disk and TOC = 1.6% continuous laser emission between 993 nm and 1092 nm with a maximum of 20.5Wat 1034 nm could be achieved. Reducing the outcoupling to 0.4% a somewhat broader tuning from 990 nm to 1109 nm was possible. With this outcoupling mirror output powers of more than 10W over a tuning range of 90 nm have been achieved. The smaller decrease of the output power around 1060 nm compared to 1034 nm can be explained by a lower inversion in the 3 at.% crystal compared to the 1 at.% crystal, which leads to a smoother gain profile. Using a HR outcoupling mirror a maximum laser emission wavelength of 1134.5 nm could be observed, but at some wavelengths above 1120 nm self-Q-switching occured, which caused a damage of the crystal and the mirror and prevented a reproducable tuning curve.
Alltogether a tuning range of Yb:Lu2O3 of 147.5 nm from 987 nm to 1134.5 nm has been achieved, which is to the best of our knowledge the broadest tuning range for an Yb-doped solid state laser reported so far.
In summary we have demonstrated for the first time the full lasing potential of Yb:Lu2O3. Employing the thin disk geometry with its multi-pass pumping scheme the output power from a 3 at.%-doped disk of 0.25mm thickness reached 32.6Wunder pumping with 45.3Wof incident power at 976 nm; the optical-to-optical efficiency was 72% and the laser operated with a slope efficiency of 80%. Thus, compared to the standard thin disk laser material Yb:YAG it was possible to increase the optical-to-optical efficiency by about 10%with Yb:Lu2O3. Additionally the higher thermal conductivity promises likewise an improvement of the beam quality in high power applications in the kW range. In order to reduce the thermal load further, thinner disks with higher doping levels will be investigated in the future.
Using a birefringent filter, high power cw tuning of the laser output from 987 nm to 1127 nm with more than 10W of output power over a tuning range of 90 nm was demonstrated.
References and links
1. A. Giesen, H. Hügel, A. Voss, K. Witting, U. Brauch, and H. Opower, “Scalable Concept for Diode-Pumped
High- Power Solid-State Lasers,” Appl.
Phys. B 58, 365–372
2. A. Giesen, “High-Power Thin-Disk Lasers,” presented at the Advanced Solid-State Photonics, Vancouver, Canada, 28–31 Jan. 2007.
3. M. Larionov, J. Gao, S. Erhard, A. Giesen, K. Contag, V. Peters, E. Mix, L. Fornasiero, K. Petermann, G. Huber, J. Aus der Au, G. J. Spühler, R. Paschotta, U. Keller, A. A. Lagatsky, A. Abdolvand, and N. V. Kuleshov, “Thin disk laser operation and spectroscopic characterization of Yb-doped sesquioxides and potassium tungstates,” OSA TOPS Vol. 50, Advanced Solid-State Lasers , pp. 625 (2001).
4. R. Peters, C. Kränkel, K. Petermann, and G. Huber, “Thin-Disk Laser Operation of Yb3+-Doped NaGd(WO4)2,” presented at the Advanced Solid-State Photonics, Vancouver, Canada, 28–31 Jan. 2007.
5. C. Kränkel, J. Johannsen, R. Peters, K. Petermann, and G. Huber, “Continuous-wave high power laser
operation and tunability of Yb:LaSc3(BO3)4
in thin disk configuration,” Appl. Phys.
B 87, 217–220
6. C. Kränkel, R. Peters, K. Petermann, and G. Huber, “High Power Operation of Yb:LuVO4 and Yb:YVO4 Crystals in the Thin-Disk Laser Setup,” presented at the Advanced Solid-State Photonics, Vancouver, Canada, 28–31 Jan. 2007.
7. L. Fornasiero, E. Mix, V. Peters, K. Petermann, and G. Huber, “New Oxide Crystals for Solid State
Lasers,” Cryst. Res. Technol. 34, 255–260,
8. K. Petermann, G. Huber, L. Fornasiero, S. Kuch, E. Mix, V. Peters, and S. A. Basun, “Rare-earth-doped
sesquioxides,” J. Lumin. 87, 973–975
9. K. Petermann, L. Fornasiero, E. Mix, and V. Peters, “High melting sesquioxides: crystal
growth, spectroscopy, and laser experiments,”
Opt. Mater. 19, 67–71
12. K. Petermann, D. Fagundes-Peters, J. Johannsen, M. Mond, V. Peters, J.J. Romero, S. Kutovoi, J. Speiser, and A. Giesen,“Highly Yb-doped oxides for thin-disc
lasers,” J. Cryst. Growth 275, 135–140
13. J. Liu, M. Rico, U. Griebner, V. Petrov, V. Peters, K. Petermann, and G. Huber, “Efficient room temperature
continuous-wave operation of an Yb3+:Sc2O3
crystal laser at 1041.6 and 1094.6 nm,”
Phys. Stat. Sol. (A) 202,
14. J. Lu, J. F. Bisson, K. Takaichi, T. Uematsu, A. Shirakawa, M. Musha, K. Ueda, H. Yagi, T. Yanagitani, and A. A. Kaminskii,
ceramic laser,” Appl. Phys. Lett. 83, 1101–1103
15. K. Takaichi, H. Yagi, A. Shirakawa, K. Ueda, S. Hosokawa, T. Yanagitani, and A. A. Kaminskii,
ceramics a novel gain material for high-power solid-state
lasers,” Phys. Stat. Sol. (A)
16. M. Tokurakawa, K. Takaichi, A. Shirakawa, K. Ueda, H. Yagi, S. Hosokawa, T. Yanagitani, and A. A. Kaminskii, “Diode-pumped mode-locked
laser,” Opt. Express 14, 12832–12838
17. A.A. Kaminskii, S.N. Bagayev, K. Ueda, K. Takaichi, A. Shirakawa, S.N. Ivanov, E.N. Khazanov, A.V. Taranov, H. Yagi, and T. Yanagitani, “New results on characterization of
highly transparent C-modification Lu2O3
nanocrystalline ceramics: room-temperature tunable CW laser action of
Yb3+ ions under LD-pumping and the propagation kinetics of
non-equilibrium acoustic phonons,” Laser
Phys. Lett. 3, 375–379
18. J. Kong, D. Y. Tang, C. C. Chan, J. Lu, K. Ueda, H. Yagi, and T. Yanagitani, “High-efficiency 1040 and 1078 nm
laser emission of a Yb:Y2O3 ceramic laser with 976 nm
diode pumping,” Opt. Lett 32, 247–249
19. M. Tokurakawa, K. Takaichi, A. Shirakawa, K. Ueda, H. Yagi, T. Yanagitani, and A. A. Kaminskii, “Diode-pumped 188-fs mode-locked
laser,” Appl. Phys. Lett. 90, 071101 (2007). [CrossRef]
20. A. A. Kaminskii, M. Akchurin, R. Gainutdinov, K. Takaichi, A. Shirakawa, H. Yagi, T. Yanagitani, and K. Ueda, “Microhardness and fracture toughness
of Y2O3- and
Y3Al5O12-based nanocrystalline laser
ceramics,” Crystallography Report 50, 869–873
21. K. Takaichi, H. Yagi, J. Lu, J. F. Bisson, A. Shirakawa, K. Ueda, T. Yanagitani, and A. A. Kaminskii, “Highly efficient continuous-wave
operation at 1030 and 1075 nm wavelengths of LD-pumped
lasers,” Appl. Phys. Lett. 84, 317–319
22. A. Shirakawa, K. Takaichi, H. Yagi, J. Bisson, J. Lu, M. Musha, K. Ueda, T. Yanagitani, T. Petrov, and A. Kaminskii, “Diode-pumped mode-locked
lasers,” Opt. Express 11, 2911–2916
23. A. Shirakawa, K. Takaichi, H. Yagi, M. Tanisho, J. F. Bisson, J. Lu, K. Ueda, T. Yanagitani, and A. A. Kaminskii, “First mode-locked ceramic laser: Femtosecond Yb3+:Y2O3 ceramic laser,” Laser Phys. 14, 1375–1381 (2004).
24. D. Fagundes-Peters, N. Martynyuk, K. Lünstedt, V. Peters, K. Petermann, G. Huber, S. Basun, V. Laguta, and A. Hofstaetter, “High quantum efficiency
YbAG-crystals,” J. Lumin. 125, 238–247
25. C. Hönninger, R. Paschotta, M. Graf, F. Morier-Genoud, G. Zhang, M. Moser, S. Biswal, J. Nees, A. Braun, G. A. Mourou, I. Johannsen, A. Giesen, W. Seeber, and U. Keller, “Ultrafast ytterbium-doped bulk
lasers and laser amplifiers,” Appl. Phys.
B 69, pp. 3–17
26. R. Peters, C. Kränkel, K. Petermann, and G. Huber, “Crystal growth by the heat exchanger method and spectroscopic characterisation of high purity Yb3+-doped Lu2O3,” submitted to the 15th International Conference on Crystal Growth, Salt Lake City, USA (2007)
27. C. Kränkel, D. Fagundes-Peters, S.T. Fredrich, J. Johannsen, M. Mond, G. Huber, M. Bernhagen, and R. Uecker, “Continuous wave laser operation of
Appl. Phys. B 79, 543–546
28. H. Kühn, S. T. Fredrich-Thornton, C. Kränkel, R. Peters, and K. Petermann, “A model for the calculation of
radiation trapping and description of the pinhole
method,” Opt. Lett. article in press